scripts: Dropped list/tuple distinction in Rbyd.fetch

Also tweaked how we fetch shrubs, adding Rbyd.fetchshrub and
Btree.fetchshrub instead of overloading the bd argument.

Oh, and also added --trunk to dbgmtree.py and dbglfs.py. Actually
_using_ --trunk isn't advised, since it will probably just result in a
corrupted filesystem, but these scripts are for accessing things that
aren't normally allowed anyways.

The reason for dropping the list/tuple distinction is because it was a
big ugly hack, unpythonic, and likely to catch users (and myself) by
surprise. Now, Rbyd.fetch and friends always require separate
block/trunk arguments, and the exercise of deciding which trunk to use
is left up to the caller.
This commit is contained in:
Christopher Haster
2025-03-31 18:09:49 -05:00
parent 1ac3aae92b
commit 82f4fd3c0f
8 changed files with 350 additions and 361 deletions
+73 -65
View File
@@ -90,10 +90,10 @@ def bdgeom(s):
return int(s, b)
# parse some rbyd addr encodings
# 0xa -> [0xa]
# 0xa.c -> [(0xa, 0xc)]
# 0x{a,b} -> [0xa, 0xb]
# 0x{a,b}.c -> [(0xa, 0xc), (0xb, 0xc)]
# 0xa -> (0xa,)
# 0xa.c -> ((0xa, 0xc),)
# 0x{a,b} -> (0xa, 0xb)
# 0x{a,b}.c -> ((0xa, 0xc), (0xb, 0xc))
def rbydaddr(s):
s = s.strip()
b = 10
@@ -124,7 +124,7 @@ def rbydaddr(s):
else:
addr.append(int(s, b))
return addr
return tuple(addr)
def crc32c(data, crc=0):
crc ^= 0xffffffff
@@ -439,9 +439,9 @@ class Rattr:
self.tag = tag
self.weight = weight
if isinstance(blocks, int):
self.blocks = [blocks]
self.blocks = (blocks,)
else:
self.blocks = list(blocks)
self.blocks = blocks
self.toff = toff
self.tdata = tdata
self.data = data
@@ -515,9 +515,9 @@ class Ralt:
self.tag = tag
self.weight = weight
if isinstance(blocks, int):
self.blocks = [blocks]
self.blocks = (blocks,)
else:
self.blocks = list(blocks)
self.blocks = blocks
self.toff = toff
self.tdata = tdata
self.jump = jump
@@ -570,9 +570,9 @@ class Rbyd:
gcksumdelta=None,
corrupt=False):
if isinstance(blocks, int):
self.blocks = [blocks]
self.blocks = (blocks,)
else:
self.blocks = list(blocks)
self.blocks = blocks
self.trunk = trunk
self.weight = weight
self.rev = rev
@@ -615,51 +615,7 @@ class Rbyd:
return hash((frozenset(self.blocks), self.trunk))
@classmethod
def fetch(cls, bd, blocks, trunk=None):
# multiple blocks? unfortunately this must be a list
if isinstance(blocks, list):
# fetch all blocks
rbyds = [cls.fetch(bd, block, trunk) for block in blocks]
# determine most recent revision
i = 0
for i_, rbyd in enumerate(rbyds):
# compare with sequence arithmetic
if rbyd and (
not rbyds[i]
or not ((rbyd.rev - rbyds[i].rev) & 0x80000000)
or (rbyd.rev == rbyds[i].rev
and rbyd.trunk > rbyds[i].trunk)):
i = i_
# keep track of the other blocks
rbyd = rbyds[i]
rbyd.blocks += tuple(
rbyds[(i+1+j) % len(rbyds)].block
for j in range(len(rbyds)-1))
# and patch the gcksumdelta if we have one
if rbyd.gcksumdelta is not None:
rbyd.gcksumdelta.blocks = rbyd.blocks
return rbyd
block = blocks
# blocks may also encode trunks
block, trunk = (
block[0] if isinstance(block, tuple)
else block,
trunk if trunk is not None
else block[1] if isinstance(block, tuple)
else None)
# bd can be either a bd reference or a preread block
#
# preread blocks can be useful for avoiding race conditions
# with cksums and shrubs
if isinstance(bd, Bd):
# seek/read the block
data = bd.readblock(block)
else:
data = bd
def _fetch(cls, data, block, trunk=None):
# fetch the rbyd
rev = fromle32(data[0:4])
cksum = 0
@@ -756,6 +712,39 @@ class Rbyd:
gcksumdelta=gcksumdelta,
corrupt=not trunk_)
@classmethod
def fetch(cls, bd, blocks, trunk=None):
# multiple blocks?
if not isinstance(blocks, int):
# fetch all blocks
rbyds = [cls.fetch(bd, block, trunk) for block in blocks]
# determine most recent revision
i = 0
for i_, rbyd in enumerate(rbyds):
# compare with sequence arithmetic
if rbyd and (
not rbyds[i]
or not ((rbyd.rev - rbyds[i].rev) & 0x80000000)
or (rbyd.rev == rbyds[i].rev
and rbyd.trunk > rbyds[i].trunk)):
i = i_
# keep track of the other blocks
rbyd = rbyds[i]
rbyd.blocks += tuple(
rbyds[(i+1+j) % len(rbyds)].block
for j in range(len(rbyds)-1))
# and patch the gcksumdelta if we have one
if rbyd.gcksumdelta is not None:
rbyd.gcksumdelta.blocks = rbyd.blocks
return rbyd
# seek/read the block
block = blocks
data = bd.readblock(block)
# fetch the rbyd
return cls._fetch(data, block, trunk)
@classmethod
def fetchck(cls, bd, blocks, trunk, weight, cksum):
# try to fetch the rbyd normally
@@ -772,6 +761,15 @@ class Rbyd:
return rbyd
@classmethod
def fetchshrub(cls, rbyd, trunk):
# steal the original rbyd's data
#
# this helps avoid race conditions with cksums and stuff
shrub = cls._fetch(rbyd.data, rbyd.block, trunk)
shrub.blocks = rbyd.blocks
return shrub
def lookupnext(self, rid, tag=None, *,
path=False):
if not self or rid >= self.weight:
@@ -1516,9 +1514,9 @@ def dbg_tree(rbyd, *,
def main(disk, blocks=None, *,
trunk=None,
block_size=None,
block_count=None,
trunk=None,
color='auto',
**args):
# figure out what color should be
@@ -1536,9 +1534,19 @@ def main(disk, blocks=None, *,
block_count = block_count_
# flatten blocks, default to block 0
if not blocks:
blocks = [(0,)]
blocks = [block for blocks_ in blocks for block in blocks_]
blocks = list(it.chain.from_iterable(blocks)) if blocks else [0]
# blocks may also encode trunks
blocks, trunk = (
[block[0] if isinstance(block, tuple)
else block
for block in blocks],
trunk if trunk is not None
else ft.reduce(
lambda x, y: y,
(block[1] for block in blocks
if isinstance(block, tuple)),
None))
with open(disk, 'rb') as f:
# if block_size is omitted, assume the block device is one big block
@@ -1548,7 +1556,7 @@ def main(disk, blocks=None, *,
# fetch the rbyd
bd = Bd(f, block_size, block_count)
rbyd = Rbyd.fetch(bd, blocks)
rbyd = Rbyd.fetch(bd, blocks, trunk)
# print some information about the rbyd
print('rbyd %s w%d, rev %08x, size %d, cksum %08x' % (
@@ -1587,6 +1595,10 @@ if __name__ == "__main__":
nargs='*',
type=rbydaddr,
help="Block address of metadata blocks.")
parser.add_argument(
'--trunk',
type=lambda x: int(x, 0),
help="Use this offset as the trunk of the tree.")
parser.add_argument(
'-b', '--block-size',
type=bdgeom,
@@ -1595,10 +1607,6 @@ if __name__ == "__main__":
'--block-count',
type=lambda x: int(x, 0),
help="Block count in blocks.")
parser.add_argument(
'--trunk',
type=lambda x: int(x, 0),
help="Use this offset as the trunk of the tree.")
parser.add_argument(
'--color',
choices=['never', 'always', 'auto'],